Semiotics and human-robot interaction

Abstract

This paper describes a robot control architecture supported on a human-robot interaction model obtained directly from semiotics concepts. The architecture is composed by a set of objects defined after a well known model of semiotic signs. These objects form an algebraic system suitable to develop formal models of human-robot interactions. They are easily mapped into motion skills and can be associated with basic forms of meaning representation. This characteristic makes them suitable to use as basis objects in high level interactions involving mobile robots. Real and simulation experiments using unicycle robots illustrate the operation of the architecture.